WalidBenzineb/E-commerce-Data-Analysis-
Walid Benzineb
E-commerce Data Analysis Project
Table of Contents
Project Overview
This project analyzes an e-commerce dataset to uncover actionable insights for business growth and optimization. The analysis covers customer behavior, product performance, sales trends, and market basket analysis.
Dataset Description
The dataset consists of several CSV files:
orders.csv: Contains order information (orderId, accountId, orderDate, etc.)order_details.csv: Links orders to products and quantitiesproducts.csv: Product information (productId, name, category, price, etc.)users.csv: Customer demographic datatransactions.csv: Payment transaction detailsTechnical Setup
Folder Structure
e-commerce-Data-Analysis/
│
├── data/
│ ├── orders.csv
│ ├── order_details.csv
│ ├── products.csv
│ ├── users.csv
│ └── transactions.csv
│
├── scripts/
│ ├── data_preprocessing.py
│ ├── rfm_analysis.py
│ ├── cohort_analysis.py
│ ├── product_analysis.py
│ ├── sales_analysis.py
│ └── main.py
│
├── outputs/
│ ├── figures/
│ └── reports/
│
├── requirements.txt
└── README.md
Copy
Script Descriptions
data_preprocessing.py: rfm_analysis.py: cohort_analysis.py: product_analysis.py: sales_analysis.py: main.py: How to Use
Clone the repository
Install required packages: pip install -r requirements.txt
Extract the CSV data files in the
data/ directory.Run the main analysis script: python scripts/main.py
View the results in the
outputs/ directory: Requirements
The project requires the following Python packages:
pandas
matplotlib
seaborn
scikit-learn
mlxtend
Analysis and Insights
Customer Demographics and Segmentation
Graph Analysis:
Age Distribution:
Right-skewed distribution with a significant peak in the 65+ age group.
Notable dip in the 18-24 age group, followed by a rise in the 25-34 group.
Relatively consistent customer numbers in the 35-64 age ranges.
Customer Segments:
"Average Customers" form the largest segment, followed by "Lost Customers".
"Best Customers" and "New Customers" are smaller but crucial segments.
Key Insights and Actions:
Develop an "Easy Shop" interface for the 65+ demographic, emphasizing accessibility features.
Create targeted campaigns for the 25-34 age group, focusing on mobile and social media integration.
Implement a stratified loyalty program to move customers from "Average" to "Best" segments.
RFM Analysis and Customer Retention
Graph Analysis:
RFM Distribution:
Cohort Analysis:
Key Insights and Actions:
Implement a "Second Purchase" campaign targeting customers 150 days after their first purchase.
Develop a VIP program for high monetary value customers (top 10% of spenders).
Create a re-engagement series for customers approaching the 180-day mark since last purchase.
Product Performance and Cross-Selling
Graph Analysis:
-Top Product Categories:
Relatively even distribution among the top 10 categories.
Tuna, Ball, and Chips are the top three categories, each with sales just over $1.2 million.
Gradual decrease in sales from the top to the 10th category, but all exceed $1 million in sales.
Association Rules:
Key Insights and Actions:
Sales Trends and Payment Preferences
Graph Analysis:
Monthly Sales Trend:
Payment Methods Distribution:
Key Insights and Actions:
Impementation Plan
Q3 2024:
Launch the "Easy Shop" interface for 65+ demographic
Develop and implement the stratified loyalty program
Q4 2024:
Create and launch the "Second Purchase" campaign
Prepare for the January 2025 promotional campaign
Q1 2025:
Implement the cross-category bundling strategy
Roll out the recommendation system based on association rules
Q2 2025:
Introduce new payment incentives to balance payment methods
Launch the re-engagement series for at-risk customers
Conclusion
This comprehensive analysis provides a data-driven foundation for strategic decision-making in our e-commerce business. By leveraging these insights and implementing the suggested actions, we can expect improvements in customer retention, average order value, and overall sales performance. The technical approach demonstrated here ensures that our strategies are based on robust data analysis and can be continually refined as new data becomes available.
Contact Information
Walid Benzineb - benzinebwal@gmail.com
Like this project
Posted Aug 6, 2024
This project analyzes an e-commerce dataset to uncover actionable insights for business growth and optimization.
Likes
0
Views
0
